Typically, when a user is interested in an object, the user manually takes a picture of the object and sends the picture to a backend system for analysis. This requires the user to aim, select a capture picture button, then send the captured picture to the backend system to perform an image search for generic or static information. The results are then typically returned on a new window that is not associated with the captured image. If the search results are not correct, the user may need to capture another picture of the object and send it back to the backend system for a further search.